Seawater Feasibility

7/26/2019 Seawater Feasibility

1/22

PAPER NO: TP03-17

CATEGORY: MECHANICAL DRAFT TOWERS

COOLING TECHNOLOGY INSTITUTE

FEASIBILITY OF SEAWATER COOLINGTOWERS FOR LARGE-SCALE

PETROCHEMICAL DEVELOPMENT

DR. SHAHRIAR EFTEKHARZADEHDR. MUIN M. BAASIRI

BECHTEL CORPORATION

PAUL LINDAHL, JR.

MARLEY COOLING TECHNOLOGIES

The studies and conclusions reported in this paper are the results of the authors own work. CTI has not investigated, and CTI

expressly disclaims any duty to investigate, any product, service process, procedure, design, or the like that may be describedherein. The appearance of any technical data, editorial material, or advertisement in this publication does not constituteendorsement, warranty, or guarantee by CTI of any product, service process, procedure, design, or t he like. CTI does notwarranty that the information in this publication is free of errors, and CTI does not necessarily agree with any statement or opinionin this publication. The user assumes the entire risk of the use of any information in this publication. Copyright 2003. All rightsreserved. This paper has been reviewed by members of the Cooling Technology Institute and approved as a valuable contributiono cooling tower literature; and presented by the author at the Annual Conference of CTI.

Presented at the 2003 Cooling Technology Institute Annual ConferenceSan Antonio, Texas February 10-13, 2003


2/22

Feasibility of Seawater Cooling Towers for

Large-Scale Petrochemical Development

Shahriar Eftekharzadeh1, Muin M. Baasiri

2, and Paul A. Lindahl Jr.

3

ABSTRACT

The major feasibility issues concerning the applicability of Mechanical DraftSeawater Cooling Towers for a large-scale petrochemical development in the hot-humid Arabian Peninsula (Gulf) region were investigated.

The main issue addressed was the impact of salts in the cooling water as it affectsthe thermal performance, permissible salts concentration, salts emission (drift) and

potential impacts upon the environment. Also addressed were tower system designmodifications to suit the Gulf region extreme weather conditions, the operation &maintenance concerns, and life-cycle costs.

It was found that seawater cooling towers, when properly designed and managed,could satisfy the cooling needs of large-scale petrochemical development in theGulf region, without significant problems. Cost of cooling was found to be lessthan the conventional once-through system normally used.

Based on the findings of this study, plans are underway to double the size of one ofthe largest petrochemical complexes in the world, located in the region.

Introduction

Petrochemical industries generate large quantities of waste heat. Therefore,cooling system considerations constitute an important aspect of petrochemical

development schemes. Technically feasible and economically favorable industrialcooling solutions are essential.

This study was performed for a major petrochemical complex located on the North-Eastern shores of the Arabian Peninsula. The complex has been operating asuccessful Once-Through central seawater cooling system since 1982. Currently,an average flow of about 650,000 m3/hr of seawater is supplied to 17 primary

petrochemical industries, and returned to the Gulf with a maximum temperature riseof 10 oC. Industries are charged a flat rate of $13.6 per 1000 m3 of seawater use forthis service. The existing system has an ultimate capacity of 1,040,000 m3/hr,which is expected to be reached by the year 2006.

1 PhD, PE, Hydraulic Engineering Specialist and Project Engineer, Bechtel Corporation, JubailProject, Jubail Industrial City, Saudi Arabia

2 PhD, PE, Principal and Manager of Engineering, Bechtel Corporation, Jubail Project, JubailIndustrial City, Saudi Arabia

3 Director, Marketing & Business Development, Marley Cooling Technologies, Inc., Overland Park,Kansas, USA


3/22

Feasibility of Seawater Cooling Towers for Large-Scale Petrochemical Development

2

Currently a seawater demand of 1,300,000 m3/hr in the existing City is projected byyear 2010. In addition, a second adjacent industrial park, with a projected heatload comparable to the existing City is being developed. To enable such expansion,a technically and economically feasible industrial cooling scheme had to be

identified. Economic and technical studies were conducted on the feasibility ofvarious cooling alternatives were undertaken. It was concluded that MechanicalDraft Seawater Cooling Towers were the most feasible alternative.

This paper summarizes the major feasibility aspects of seawater cooling towers thatwere investigated for this petrochemical complex. Its findings are applicable to

petrochemical units and other major waste heat producing plants in the Gulf region.

Once-Through Versus Recirculating System

Cooling systems are either Once-through or Recirculating. A Once-throughsystem uses the cooling water only once before it is discharged. A Recirculatingsystem recycles the cooling water after it has been cooled at a heat sink.

Cooling systems are also classified as Closed or Open. In a Closed system, thesink is a heat exchanger, and the system is not exposed to the atmosphere. In anOpen system, the sink is either an evaporative cooling facility (such as a pond or acooling tower), or a free water body (such as a lake, river, or sea), such that thesystem is open to the atmosphere.

Figure 1: Typical Once-through system

Figure 1 represents the seawater cooling arrangement used in the petrochemicalcomplex of this study. It is an arrangement commonly used in the Petrochemical

Return

Heat

Exchange

Industry

Heat

Exchange

Makeup

(Losses)

Supply

Utility

CClloosseedd

RReecciirrccuullaattiinngg

--OOnnccee

tthhrroouugghh

SEAWATER FRESH WATER


4/22


3

complexes in the Gulf region. It is a Once-through system (operated by the utilitycompany), coupled with a fresh-water Closed Recirculating system (operated byindividual industries). The utility department distributes the entire cooling flow toindividual industries and receives the return flow for discharge.

The Once-through portion of the above arrangement may be replaced by coolingtowers in an Open Recirculating system. The industry portion (ClosedRecirculating part) would remain mostly unchanged. Figure 2 shows the proposedsystem with cooling towers.

Figure 2: Proposed system with cooling towers

Both the Open Recirculating (Cooling Towers), and the Closed Recirculatingsystems would be operated by the industry. The role of the utility company would

be to provide the Makeup water required, and to receive the Blowdown from thetowers. This means significant reductions in the scope and cost of the water Supplyinfrastructure (normally a major seawater pumpstation), as well as the Distribution

and the Return systems (canals or large-diameter piping) that must be furnished bythe utility company. This is because Makeup flow is only about 6%, andBlowdown is only about 4% of the cooling water flow. This constitutes a majoradvantage of this system over the Once-through system as it impacts the firstinvestment costs, and future expansion possibilities.

Evaporation

Drift

Acid

Anti ScalantHeat

ExchangeBlowdown

Chlorine

Cooling

Tower

Industry

Heat

Exchange

Makeup

Utility

OOppeennRReecciirrccuullaattiinngg

FRESH WATERSEAWATER

Makeup(Losses)

CClloosseedd

RReecciirrccuullaattiinngg


5/22


4

For a petrochemical complex already using the Once-through system, industries canuse the above cooling tower arrangement either as a substitute or as a supplement tothe Once-through system, since it does not impact their closed loop system.

One alternative with existing Once-through systems is to use cooling towers on the

Return line as helper towers. This can nearly double the availability of seawater.No makeup or blowdown is necessary as the system is not Recirculating.

Salts in the Cooling Water

What differentiates seawater cooling towers from fresh water towers is theexistence of dissolved minerals (salts) in the cooling water. Therefore, establishingthe impact of salts in the cooling water is the single most important technicalfeasibility concern. The areas of concern were identified as thermal performance,salts concentration, salts emission (Drift) and environmental impacts, and O&M.

Thermal Performance

Salt in the water has four basic effects on its use as a coolant, only one of which ismajor. Salt lowers the vapor pressure of water, thus the water does not evaporate asreadily. This makes it less as a effective coolant and reduces tower performance (1).Table 1 shows an example of the effect of salts in water upon vapor pressure (4).

Table 1: Impact of salts in water upon vapor pressure (4)

Physical property Fresh Water Seawater1)

Water Temp (oC) 35 35

Air Temp (oC) 30.6 30.6

Air Relative Humidity (%) 60 60

Liquid Vapor Pressure (kPa) 5.62 5.42

Air Vapor Pressure (kPa) 2.63 2.63

Liquid-Air Vapor Pressure Difference (kPa) 2.99 2.79

Liquid-Air Vapor Pressure Difference(% of Fresh Water Condition)

100 93.2

Performance loss (%) 2) - 5.4

1) At salts concentration of 50,000 ppm

2) Performance loss (approximated as 80% of change in VP difference) = 0.8x (1-0.932)

The Research and Development Division of Fluor Corporation (11) summarized

corrosion and thermal performance results conducted on fresh water, and water withsalt levels of 34,000 and 62,000 ppm. Their conclusion was that increasing thedesign wet bulb temperature by 0.055 oC per 4,000 ppm TDS was a satisfactorycompensation for the salt effect on cooling tower performance (1).

Cooling tower vendors recommend degrading the tower performance byapproximately 1.1% for every 10,000 ppm of salts in the cooling water. In


6/22


5

practice, most engineering contractors specify a 0.55-1.1 oC margin on the wet bulbtemperature to account for salts in the cooling water (2).

Salts also increase density and surface tension, while decreasing specific heatcapacity. The increase in density has a small positive effects on performance,

while lower heat capacity and the higher surface tension have a negative effect.The net impact is a small negative effect. Table 2 compares the density and specificheat capacity of fresh water and seawater at 50,000 ppm salts concentration.

Table 2: Effect of salts on water density and specific heat capacity (4)

Physical property 1)

Fresh Water Seawater 2)

Difference (%)

Density (kg/m3) 989.9 1026.8 + 3.7

Specific HeatCapacity (kJ/kg-oC)

4.178 3.952 - 5.4

1) Corresponding to temperature at 49oC2) At salts concentration of 50,000 ppm

Therefore, the heat capacity decreases more than the density increases. Since heattransfer is proportional to the product of Density and Specific Heat capacity, the neteffect is slightly less heat absorbing capacity of seawater as compared with freshwater. However, this effect is minor as compared with the vapor pressure effect.

It follows that the net impact of salts in the cooling water is that it reduces theeffective Approach (defined as the difference between wet bulb temperature andcold water temperature). For design purposes, a maximum value of 1.1 oCreduction in Approach would be used. This impacts the tower size required to

achieve the same cold water temperature. The impact of Approach upon tower sizeusing fresh water is shown in Figure 3 (3).

Figure 3: Impact of Approach on tower size using fresh water (3)


7/22


6

The shape of the curve in Figure 3 means that the impact of reduced Approach upontower size is more pronounced at the lower end of the Approach scale. With 2.8oC (5.0 oF) being considered as the lowest practical Approach for fresh water, thecorresponding minimum Approach for salt water could shift to as high as 3.9 oC

(2.8

o

C limit + 1.1

o

C allowance for salts). Consequently, the tower size could beup to 20% higher i.e. the worst practical impact of salts in the cooling water is torequire 20% more tower size (20% more cells). In practice, the actual impactdepends on the TDS in the seawater source and the concentration cycle (factor bywhich that water is permitted to concentrate). This is elaborated upon below.

Salts Concentration

Salts in the cooling water provide the potential for scaling i.e. deposition of salts inthe cooling system. The potential for scaling increases with salts concentration andis affected by the composition of the salts and the water chemistry.

The permissible concentration of salts in the cooling water is determined from the

results of the water chemical and physical analysis. For scaling potential, theconcentration of Ca++ and Mg++, and levels of pH and the alkalinity are particularlyimportant. Table 3 shows the seawater composition for this study.

Table 3: Seawater analysis results for this studyTemperature 15 oC 35 oC

Density 1.027 kg/l at 25 oC

Turbidity 1.0 2.5 NTU

Total Suspended Solids 10 30 mg/l

Conductivity 59,000-61,000 umho/cm

TDS 41,654-43,066 mg/l

pH at 25 oC 8.1 8.3

M - Alkalinity 130 - 135 mg/l as CaCO3Total Hardness 7625 - 7750 mg/l as CaCO3Ca++ 470-500 mg/l

Mg++ 1563 - 1585 mg/l

Na+ 13000 - 13165 mg/l

K+ 458 - 473 mg/l

Cl- 23,000 24,500 mg/l

SO4-- 3300 3500 mg/l

Total Fe--- 10- 12 ppb

Cu-- 6 ppb

DO2 4.4 7.0 mg/l

For the seawater composition shown in Table 3, the water treatment requirementswas sought from a reputable water treatment company. They recommendedchemical additives to control alkalinity, scaling, corrosion, and bio-fouling, as wellas set limits on the suspended solids and minimum flow velocity.


8/22


7

Most significant, was a limit on the Concentration Cycle of 1.4 (maximum TDS of60,000 ppm), which sets the quantity of makeup flow required, and a maximumWater Temperature of 49 oC, which limits the Range. The following shows howthe concentration cycle is related to the makeup flow.

The Recirculating system is operated to maintain two mass balances; 1) WaterBalance, and 2) Salts Balance. For water balance we have:

QMU = QBD+ QE+ QD (1)

Where QMU, QBD, QE, and QD are the Makeup, Blowdown, Evaporation, and Driftflow rates respectively. For salts balance we have:

CMUQMU = CCW(QBD+QD) (2)

Where CMU and CCW are salts concentration in the Makeup and Cooling Water(water being recirculated) respectively. Concentration Cycle, RC, is the ratio of CCW

to CMUand is given by rearranging Eq. 2:

RC= QMU/ (QBD+ QD) (3)

Substituting for QMUfrom (1) into (3) and rearranging gives:

QBD= (QE/ (RC-1) ) - QD (4)

Equation 4 is used to determine the Blowdown for a given Concentration Cycle,Evaporation and Drift. Blowdown is then used in Eq. 3 to determine the Makeup.At the limit of RC= 1 in Eq. 4, QBD= QMU, and the system is Once-through.

Figure 4: Variation of Makeup Flow with Concentration Cycle

Using the above procedure, makeup flow for concentration cycles ranging from 1.2to 2.0 were calculated for an example case of 100,000 m3/hr cooling flow. The


9/22


8

results are plotted in Figure 4. A maximum concentration cycle of 1.4 means that ata minimum, 5900 m3/hr of Makeup, which is 5.9 % of the cooling flow, is required.The corresponding Blowdown, as calculated from Eq. 4 is 4200 m3/hr.

In addition to scaling considerations, salts concentration must be limited to manage

corrosion per materials specifications in the cooling system. Another importantconsideration is concerning health issues. Table 4 shows a manufacturersguidelines on salt concentration limits based on materials specifications. Also

provided are limitations imposed by health considerations.

Table 4: Salt concentration limits in cooling water because of materials

specifications and health considerations (4)


10/22


9

The points noted in this section highlight the importance of the composition of thesalts in the water so as to ensure the correct limit on salts concentrations based onthe different governing considerations.

Salts Emission (Drift)

A significant reservation with using seawater cooling towers was salts emissionbrought about by the Drift, and its impact upon the environment. In fact, this hashistorically been the single reason for rejecting this technology.

Recent advances in Drift Eliminators developed for seawater cooling towers havesignificantly minimized this problem. Major manufacturers are now able toguarantee a maximum drift of 0.0005% of total cooling flow for salt waterapplications (4, 10). This drift rate is in fact the smallest amount that can be

physically measured per existing test equipment (4).

Verification of the drift rate can be required by the client as a part of the acceptancetesting. The Cooling Technology Institute (CTI) test codes ATC-140 Drift Testing

of Wet, Wet/Dry, Closed Circuit Cooling Towers is specifically designed for thispurpose and is performed by licensed agencies (4).

The 0.0005% drift converts into a small concentration of salt particles in the coolingtower plume. For a typical cooling tower cell with 5000 m3/hr having a fandiameter of 8.5 m and an air flow of about 800 m3/s (air flow velocity of about 13.9m/s), a drift of 0.0005% means a saltwater discharge of 25 liter per hour into theatmosphere. If this saltwater has a concentration of 60,000 ppm (mg/l), it means asalts discharge of 417 milligrams per second with the plume. This salt is mixed inwith the fan air flow of 800 m3 per second exhausted from the tower. Therefore,the concentration of salts in the plume is 0.52 micro grams per liter of air. Such a

concentration is several times lower than the concentration of salts in the sea-air,which originate from the seawater spray and aerosol.

An import aspect is the size of the water droplets in drift. Larger droplets result inlarger salt particulates when dry, and are more likely to fall out. Table 6 shows the

predicted mass distribution of drift particle size from modern drift eliminators (4).

Table 5: Distribution of drift particle size from modern drift eliminators (4)

Mass (%) Droplet Size (Microns)1.0 above 2754.0 230 - 275

5.0 170 - 23010.0 115 - 17020.0 65 - 11520.0 35 - 6520.0 15 - 3520.0 Below 15


11/22


10

A droplet size of about 150 microns is a reasonable cut off size for salts fallout (4).Table 5 shows that more than 90% of the droplets are smaller than 150 microns i.e.less than 10% of the drift could result in fallout. US EPA permits the use of theentire cooling tower drift as PM-10 Particulate Mass Emissions i.e. particles that

are or dry to less than 10 microns, and may therefore be treated as particulates(which remain suspended in the air).

Therefore salts deposition, near and/or down-wind of cooling towers, should not besignificant. There is comprehensive practical experience to support this. For thisstudy, three major cooling tower manufacturers provided clients lists of seawatercooling tower users. Those contacted did not report any particular problemsassociated with the drift. In fact, the drift phenomenon was unknown to someoperators. At least two long-time operating saltwater cooling tower installationswere visited. These did not exhibit any visible signs of salts fallout.

Perhaps the most convincing evidence on drift is the comprehensive studies ofChalk Point Cooling Tower Project funded by the Maryland Department of Naturalresources in the USA, from 1972 to 1980 (Refs. 5 through to 9). The studiesconcentrated on the drift from a 60,000 m3/hr (260,000 gpm) cooling tower system,circulating water at 14,000 ppm having a drift rate of 0.002% (i.e. 4 times morethan latest technology cooling towers @ 0.0005%). The studies found nomeasurable increase in soil salts concentration in the tobacco fields near the facility.

Based on the above, it was concluded that drift as it impacts salt concentration inthe air, and salts fallout from cooling towers should not pose a serious problem. Theadditional salts introduced are not expected to have a significant impact on theextreme corrosive environment of the Gulf region. Therefore, no additionalcorrosion protection measures for nearby facilities would be required, and the

specifications and measures already in use would be sufficient.Siting Issues

Siting is the location and orientation of the towers within the site and with respect toother cooling towers. Given the minimized impact of drift, siting issues withseawater cooling towers should really be no different from those with freshwatertowers. However, some additional precautionary considerations would be prudent.

The primary concern with siting is thermal performance as affected byRecirculation (entry of towers own plume into tower its intake), andInterference (entry of the plume from one tower into another). Recirculation isminimized by the correct orientation of towers with respect to the predominant

wind direction (broad side parallel to wind direction), and Interference is addressedby the correct arrangement of the towers to observe a certain minimum distancebetween towers depending on the windrose (3).

Siting requirements favor locating the cooling towers on individual industry sites.This precludes a central cooling tower arrangement for the industrial complex,which would physically concentrate the towers. A de-centralized tower


12/22


11

arrangement is also favored by considerations of cooling water distribution. Acentral cooling tower arrangement would distribute the entire cooling water flow asopposed to the Makeup flow (which is about 5%). Another advantage is the abilityto cater to individual industry needs.

Although drift is minimized, prudence still calls for special care with the siting ofseawater cooling towers. There is a potential for impact on the down-windfacilities, and a compounded effect on Interference and Recirculation (due to saltsin the plume). As a precautionary measure sensitive facilities, such as electricalswitch-gear should not be located downwind of cooling towers in general, andseawater cooling towers in particular..

Therefore, cooling towers should be sited at each industry with broad end parallel tothe predominant wind direction, as far apart from one another as possible, and awayfrom sensitive equipment (as a precaution). Any additional siting considerationsfor seawater cooling towers as compared with freshwater towers are of

precautionary nature only.

Operation and Maintenance

The operation and maintenance concerns with seawater cooling towers were rootedin doubting the suitability of cooling towers as a whole (sea or freshwater) for

petrochemical application. The concerns were compounded by the perceivedpotential complications with the use of seawater in the towers.

The main areas of concern were; performance during extreme wet bulb conditions(as it impacts the cold and hot water temperatures), performance compared with theexiting once-through system, and operation and maintenance issues (down times,redundancy requirements, clogging, chemical additives, impact of blowdown

discharge, etc.).

Performance During Extreme Wet Bulb Conditions

Engineered solutions are seldom designed for the worst possible natural event. So,there are times when and event exceeds the design conditions. For petrochemicalapplications, a stringent 1% wet bulb exceedence level was considered appropriate.The concern was cooling tower performance during exceedence times, particularlythe cold and hot water temperature, and the risk of plant shutdown.

To answer the above, a generic case of 70,000 m3/hr cooling tower system with aduty of 32 oC Design (entering) Wet Bulb, 35 oC Cold Water, and 45 oC Hot Watertemperatures, was simulated (10) for a short-spell hike in the actual wet bulbtemperature to 34.5 oC. The results predicted an increase in the cold and hot watertemperatures to 37.1 and 47.1 oC respectively, for uninterrupted industry operation.So, if an increase of 2.1 oC in both the cold and hot water temperature isacceptable, there would be no impact on the industry during such an event.

If not, one option is to increase the cooling circulation rate in order to reduceRange. Cooling towers can accommodate up to 20% increase in flow rate above


13/22


12

design (10). For such a case, thermal modeling showed that a cold and hot watertemperatures of 37.7 oC, and 46.1 oC (8.4 oC Range) respectively. Industriessensitive to the higher hot water temperature that must maintain 100% load duringsuch conditions could use this option.

If the increase in the hot water temperature is still unacceptable (because of aparticular temperature sensitive process), then the industry must either reduceproduction or design for the 34.5 oC wet bulb temperature. In the case of the latter,a 91,000 m3/hr cooling tower system with a duty of 34.5 oC WBT, 37.3 oC CWT,and 45 oC HWT (2.8 oC Approach and 7.7 oC Range) would be specified. Thiswould cost more. The same option can be used for even higher design wet bulbtemperatures (if necessary).

Assuming that the prediction models are reasonably accurate, cooling towers wouldperform adequately during extreme conditions with a number of options possible.No plant shut down is necessary. Extreme wet bulb condition usually occur during acertain period in the year (the month of August). Industry should schedule theirannual plant maintenance during this period to reduce the occurrences of operatingabove design conditions.

Performance compared with once-through system

The industrial complex of this study has been operating a central once-throughsystem for about 20 years. To be accepted, the proposed seawater cooling towerswere expected to perform as good as or better than the existing system. The notionwas that once-through systems are not exposed to environmental extremes, whereascooling towers are at the mercy of ambient conditions.

Investigation of the existing once-through system showed that industries do have to

adjust their operation to cope with the seasonal variation of Gulf water temperature.They adjust the Range and water use accordingly to limit the hot water temperaturewhile maintaining the same heat load. This is demonstrated in Figure 5.

The cooling water supply temperature increases to a high of about 35 oC in theSummer and Low of about 15 oC in the winter. To limit the return watertemperature, the collective complex response is to reduce the range to about 7 oC inthe Summer and 9.5 oC in the winter. Consequently, the water use increase to ahigh of about 15.5 M m3/day in the summer and 10.5 M m3/day in the winter.

The cooling tower system is also faced with seasonal ambient variations. It has tocope with the variations in the wet bulb temperature, which directly affect the coldwater temperature. The temperature of the cold water emerging from the tower canonly is about 3.0 oC higher than the wet bulb temperature.


14/22


13

Figure 5: The existing once-through system temperatures and water use

Y2001

Figure 8 shows the hourly ambient wet bulb temperature data for year 1999 (one ofthe hot years on record), for the site of this study. The general trend is similar tothat of the seawater temperature shown in Figure 5. The highest recorded singlehour ambient wet bulb temperature value is about 34.0 oC. The correspondingentering wet bulb temperature could be a couple of degrees higher due to the salteffect, recirculation, and interference described earlier.

J F M A M J J A S O N D

Discharge

Intake

RangeoC


15/22


14

Figure 8: Hourly ambient wet bulb temperatures Y1999

The consequence is that for short spells during the month of August the coolingtower could be faced with an entering wet bulb temperature of about 36.0 oC, andthat the coldest water temperature emerging from the tower could be as high as 39oC. This is higher than the cold water temperature of the once-through system.

Given that the limit of hot water temperature is 49 oC (see Table 4), a cold watertemperature of 39 oC still leaves 10 oC Range for the industry to utilize. Individualindustries could opt to achieve a lower hot water temperature (reduced Range) ifnecessary using options discussed in the previous section. No shut down, or evenreduced production is required.

It follows that, both the cooling tower and the once-through systems arefundamentally similar. Both are faced with seasonal variations in temperature thatdirectly impact the cold water temperature. For either system, such variations can

either be designed for initially, or coped with operationally without any significantconsequences.

Operation and Maintenance Issues

Operation and maintenance concerns were with regards to down times, redundancyrequirements, clogging, chemical additives, bio-fouling, and the environmentalimpact of blowdown discharge.

These issues were primarily investigated by contacting long-time seawater coolingtowers users. In addition, cooling tower suppliers and water treatment companieswere consulted. Tables 6 and 7 are lists of seawater and brackish-water cooling

tower installations, provided by two major tower manufacturers. Additionalinformation on operating seawater cooling tower installations may be obtained fromeach manufacturer. A recent large-scale seawater cooling tower installation is theCantarell Nitrogen Complex in Mexicos Campeche state, which processes aseawater cooling flow of 80,000 m3/hr (18).

Regarding downtimes, the message from existing seawater cooling towerinstallations was that no extensive maintenance is required. No special measuresother than an annual walkthrough inspection in each cell, is generally required.

On redundancy, there is no strict requirement to have any standby cells, as nosignificant downtime is expected. However, having one or two additional tower

cells (out of a total of 15 to 20 cells), would be a reasonable investment to ensurehigher reliability, as well as to provide for increased cooling flow capability.

Clogging problems appear to have mostly been resolved by the tower manufacturersthrough the special attention to the design of the distribution nozzles, and thedevelopment of clogging free or splash-type fill material. Acid treatment, asspecified by the water treatment company, is used to reduce the risk of scale


16/22


15

formation in the piping. Operation experience shows that these measures areapparently working.

The chemical additives used are primarily acid treatment to reduce scaling, andchlorination to control bio-fouling. The acid treatment dosage may be higher than

once-through systems because of the higher salinity, while chlorination dosage issimilar. The quantity and the cost of such treatment are reportedly not exorbitant.

The blowdown from cooling tower is more concentrated in salts than the originalseawater (by the concentration cycle). It also has acid and chlorine residuals.These must conform to environmental discharge regulations. Operation experienceshows that this can be done. The blowdown discharge from a seawater coolingtower is similar in composition to the return flow from a desalination plant, only alot less in quantity. The Gulf region accommodates large desalination plants thathave been discharging their return flow into the gulf for considerable time.

Therefore, based on the operation experience gained from existing towers, it

appears that the O&M problems associated with seawater cooling towers are for themost part manageable.


17/22


16

Table 6: List of Saltwater/Brackish Water Cooling Towers (4)

Location Owner/Project Design ConditionsFlow@HWT/CWT/WBT

1)Year

Oklahoma, USA Oklahoma G& El. Co. 13,680 m/hr@40/30/23.9C 1953

Kansas, USA American Salt Co. 1,140 m/hr@32/27.2/24C 1964

New Jersey, USA Exxon Chemical Co. 5,016 m/[email protected]/27.8/23.9C 1968

Stenungsund, Sweden ESSO Chemical AB 23,040 m/[email protected]/19/15C 1969

Judibana Falcon,Venezuela

Lagoven Amuay 7,752 m/[email protected]/33.9/29.4C 1970

Okinawa, Japan Exxon Petroleum Co. 3,329 m/[email protected]/31/27.8C 1971

Florida, USA Gulf Power Co. 37,620 m/[email protected]/32.8/28.2C 1971

Texas, USA Dow Chemical Co. 13,680 m/[email protected]/30.5/26.7C 1973

Maryland, USA Potomac El. P. Co. Plant 3 59,280 m/[email protected]/32.2/25.6C 1974

Virginia, USA Virginia Electric Co. 75,240 m/hr@45/31.7/25.6C 1975North Carolina, USA Pfizer Co. 12,442 m/[email protected]/30.6/26.7C 1975

California, USA Dow Chemical Co. 2,736 m/[email protected]/25.6/21.1C 1976

Washington, USA Italco Aluminum Co. 9,348 m/[email protected]/29.4/22.8C 1976

California, USA Pacific Gas & Electric Co. 84,816 m/[email protected]/27.8/21.1C 1976

Texas, USA Houston Light & Power Co. 54,720 m/[email protected]/34.7/27.8C 1977

Mississippi, USA Mississippi Power Co. 39,444 m/[email protected]/32.2/26.7C 1980

Maryland, USA Potomac El. Pwr. Co. Plant 4 59,280 m/[email protected]/32.2/25.6C 1981

Arizona, USA Palo Verde I Plant 133,836 m/[email protected]/30.7/25C 1985

Arizona, USA Palo Verde II Plant 133,836 m/[email protected]/30.7/25C 1986Florida, USA Stanton En. #1 Station 45,600 m/[email protected]/32.8/25.6C 1986

Table 6: List of Saltwater/Brackish Water Cooling Towers (4) Continued

Location Owner/Project Design Conditions

Flow@HWT/CWT/WBT1)

Year

Arizona, USA Palo Verde III Plant 133,836 m/[email protected]/30.7/25C 1987

Texas, USA Houston L. & Power Co. 54,948 m/[email protected]/34.7/27.8C 1987

Delaware, USA Delmarva Power & Light 46,170 m/[email protected]/32.2/26.1C 1989

California, USA Delano Biomass En. Co. 4,423 m/[email protected]/28.3/22.7C 1991

Florida, USA Stanton En. #2 Station 45,600 m/[email protected]/32.8/25.6C 19951) Hot Water/Cold Water /Wet Bulb Temperatures


18/22


17

Table 7: Installation List of Seawater Cooling Towers (10)Year Client Project Country Flow

(m3/hr)

1973 I. S. A. B. SIRACUSA IT 16,000

1973 ATLANTIC CITY ELECTRIC CO (NJ) BEESLEY'S POINT US 14,423

1976 PUBLIC SERV. ELEC. & GAS CO HOPE CREEK US 250,7601978 E. B. E. S. - DOEL NUCLEAR PP DOEL BE 183,2401979 JEDDAH INT. AIRPORT JEDDAH SA 35,4001981 JACKSONVILLE ELEC. AUTH. JACKSONVILLE (FL) US 112,520

1984 GUJARAT ELECTRICITY BOARD PANANDRA KUTCH -GUJARAT

IN 33,100

1985 SIAPE SFAX TN 8,0001990 FLORIDA POWER CORP. ST PETERSBURG US 156,0001990 C. E. G. B. KILLINGHOLME GB 46,8721991 BASF ANVERS BE 14,5001992 ATLANTIC CITY ELECTRIC CO B. L. ENGLAND, N. J. US 16,2801993 POWERGEN CONNAH'S QUAY GB 85,3921993 E. G. A. T. BANG PAKONG TH 71,1001995 E.G.A.T. SOUTH BANGKOK TH 33,5001996 AMATA EGCO B BANG PAKONG TH 12,1681996 MEDWAY POWER Ltd MEDWAY GB 35,3801997 GEM METHANOL TRINIDAD TRINIDAD 12,5131997 ECOELECTRICA, LP PENUELAS 2,1841997 ECOELECTRICA, LP PENUELAS 35,4081998 EGAT KRABI TH 48,1001999 KALTIM PARNA INDUSTRY BONTANG ID 17,0001999 ESSO SINGAPORE PVT LTD SINGAPORE SG 4,0881999 FLORIDA POWER COPR

CRYSTAL RIVER PLANTCRYSTAL RIVERFLORIDA

US 67,229

2000 ESSO SINGAPORE PTE LTD SINGAPORE SG 14,082

2000 ENDESA SAN ROQUE ES 16,1422000 ST JOHNS RIVER POWER PARK JACKSONVILLE FL. US 56,2582001 GB3 LUMUT MY 34,0502001 ENDESA TARRAGONA ES 28,2722001 PETROBRAS TERMORIO BR 55,0002002 JUBAIL UNITED

PETROCHEMICALJUBAIL SA 66,605

Costs

The true cost of a cooling system to the industry is determined by accounting forboth the initial and the running costs over the economic life of the system (life-cycle cost). The initial costs are comprised of equipment purchase, transport,customs clearance, taxes, land acquisition, power acquisition, civil, mechanical,electrical, piping works, and testing and commissioning. The operation andmaintenance costs include makeup & blowdown charges, electricity, watertreatment, O&M crew, parts, and materials.

For this study, a typical 70,000 m3/hr system with a duty of 45 oC HWT, 35oCCWT, and 32 oC WBT was selected for life-cycle cost analysis. Such a tower would


19/22


18

require a total fan power of 2.9 MW and total pump power of 4.1 MW (towerpumping only, excludes pumping through system piping and heat exchangers).Table 8 shows the estimated installed cost of such a system in Saudi Arabia.

Table 8: The installed cost of a 70,000 m3/hr seawater cooling tower in Saudi Arabia

Equipment Procurement $ 5,170,000Concrete Structure $ 2,250,000External Piping & Pumps $ 380,000Electrical Hookups $ 800,000Equipment Installation $ 520,000

Total Installed Cost $ 9,120,000

The cost of land for the 1.0 ha required, was negligible. At 8% discount rate and 15years, the total installed cost in Table 8 is equivalent to $1,065,485 per year.

Table 9 shows the basis for the annual operation and maintenance costs. Based onthe figures shown in Tables 8 and 9, as well as data obtained from water treatmentcompanies and tower manufacturers (for water treatment and routine O&M costsrespectively), the estimated Total Annual Costs were as shown in Table 10.

Table 9: Basis for estimating annual O&M costs

Cost of Electricity $ 0.032 per kW-hr,Charge for Makeup $ 13.3 per 1000 m3 of seawater *Charge for Blowdown $ 0 (included in the charge for makeup)*Makeup 30.7 Mm3/yr (at 5%, 3500 m3/hr)Pump operation time 8,760 hrs/yr,Fan operation time 7,500 hrs/yr

* Same as seawater charge for existing once-through system

Table 10: Total Annual Cost of a 70,000 m3/hr seawater cooling tower in Saudi Arabia

Capital Investment $ 1,065, 485 28.3%Electricity (Fan + Pumps) $ 1,837,705 48.9Makeup & Blowdown $ 408,800 10.9

Water Treatment $ 300,000 8.0Routine O&M $ 150,000 4.0

Total Annual Costs $ 3,761,990 per year 100 %

Table 10 shows that the electricity costs constitute almost half of the annual cost ofthe seawater cooling tower, with the capital investment comprising about one third.


20/22


19

Larger towers require less power for operation, so there may be some potential foroptimization by increasing the tower size (capital investment) and thereby reducingthe power required (electricity cost), so as to minimize the overall cost.

Makeup and Blowdown are payments for the seawater makeup-supply/blowdown-

return (Makeup) service, to the provider, and rank third at 10.9%. However, thismay vary depending on the rates charged for this service. Water treatment andO&M costs rank 4th and 5th respectively, and are relatively low.

At 70,000 m3/hr flow and 8,760 hours of pump operation, the system delivers a totalof 613.2 M m3 per year cooling seawater to the industry. Therefore the unit cost ofseawater for the industry is $6.14 per 1000 m3. This is less than half charge for theseawater-supply/hotwater-return service industries currently pay for once-throughcooling. So, seawater cooling towers would cost the industry about 50% less thanthe existing once-through system. This was a significant finding of this study.

A sensitivity analysis showed that the unit cost of $6.14 per 1000 m3 is quite robust.For example, increasing the equipment procurement cost by 50% (from $ 5,170,000to $7,755,000) would increase the unit cost by about 8% (from $6.14 to $6.63).Similarly, increasing the water treatment cost by 100% (from $300,000 to$600,000) would increase the unit cost by about the same (from $6.14 to $6.62).

A controversial issue was the charge for Makeup. It was argued that setting thisto equal the once-through charge for seawater-supply/hotwater-return, is not correctand that a charge of $133.9 per 1000 m3, which would fully recover the cost of theMakeup infrastructure for the new industrial park should be used in the analysis.

With the above rate, the annual cost of Makeup jumps to $4,104,352, and the total

annual cost becomes $7,757,542. Makeup cost is now the biggest cost at 52%,with electricity at 23.2% and capital cost at 13.7% ranking second and thirdrespectively. The unit cost of seawater increases to $12.65 per 1000 m3, which isstill less than the existing charge of $13.3 per 1000 m3 with the once-throughsystem. This was an important find, which provided the necessary reassurance to

proceed with this technology for the development of the new industrial park.

Summary and Conclusions

This study investigated the feasibility of seawater cooling towers for a majorpetrochemical complex development in the Gulf region.

Salts in cooling water affect the thermal performance, are emitted in the drift, causescaling and bio-fouling, and raise environmental and O&M concerns. The netimpact on thermal performance, is to reduce the Approach by a up to about 1.1 oC.This would require up to 20% larger tower.

Drift is minimized by drift eliminators, which are claimed to reduce the drift ratedown to 0.0005%. This converts into a small concentration of salts in the plume.


21/22


20

Tests to verify this at commissioning are available through CTI. Data on driftdroplet size shows the most droplets to be too small to fall out. Evidence fromoperating towers suggests insignificant salts fallout.

Scaling is controlled by limiting concentration cycle and acid treatment, based on

the water chemistry. Bio-fouling is controlled by chlorination in the similar dosageas for once-through systems.

This study did not directly investigate environmental impacts. However, if driftrate is indeed as low as claimed, then environmental impacts from towers emissionshould be minimal. No significant impacts from blowdown return is expected.

Numerical simulation of Tower thermal performance during extreme wet bulbconditions of 35 oC showed that the usual not-to-exceed hot water temperature of45 oC can be maintained. For the site of this study, a design entering WBT of 32 oCis satisfactory, and a maximum hot water temperature of 49 oC may be tolerated.

Existing seawater users were contacted on O&M issues. No major O&M problems

that could be considered difficult, expensive, or objectionable were reported.

Life-cycle cost analysis showed that the unit cost of water from cooling towers isabout 50% cheaper than the existing once-through system.

Based on the findings of this study, it may be concluded that:

1- Using seawater in cooling towers is technically and economically feasible.

2- Seawater cooling towers can satisfy the cooling needs of petrochemicalindustries. WBT variations are handled by design or operation.

3- No significant O&M complications with seawater cooling towers is reported.

4- Seawater cooling tower economics are favorable as compared with the once-

through system.5- Petrochemical development schemes or individual plants planning expansions

would be wise to consider seawater cooling towers as an alternative.

Based on the above conclusions plans are under way to double the size of what isalready the largest petrochemical industrial complex in the region.

References

1- Hamilton, Thomas H., 2001. Exerts from Report on Seawater Cooling Towersprepared for the Client under contract.

2- Foster Wheeler Process Plants Division, Process Standard, Heat Transfer,

Cooling Tower and Systems

3- Marley Cooling Tower, 1998 Cooling Tower Fundamentals, The MarleyCooling Tower Company.

4- Paul Lindahl, 2001 Cooling Towers for Jubail, presentation made to theRoyal Commission for Jubail and Yanbu. Jubail, Saudi Arabia, 13 June,


22/22

21

5- Chalk Point Cooling Tower Project, 1977, Volume 1, Salt Loading, Modeling& Aircraft Hazard Studies, NTIS, US Dept. of Commerce, August.

6- Chalk Point Cooling Tower Project, 1978, Cooling Tower Effects on Cropsand Soils. Post Op. Report No. 3, NTIS, US Dept. of Commerce, July.



9- Chalk Point Cooling Tower Project, 1979, Drift Salinity Experiments,NTIS, US Dept. of Commerce, October.

10- Hamon Group, Hamon Industrie Thermique, 116/118, rue Jules Guesde B.P.77, 92303 LEVALLOIS PERRET CEDEX, France, Tel: (33) 1 45 19 36 36

11- Impact of Salinity on Corrosion - The Research and Development Division,

Fluor Corporation, January 195712- Marley Cooling Tower Co., 7401 W. 129 St., Overland Park, KS 66313, USA

13- Corrosion Engineering Manual, Corrosion Control Department, , RoyalCommission for Jubail and Yanbu, 25 Rajab 1415H (27 December 1994).

14- Sereda, Weather Factors Affecting the Corrosion of Metals, Corrosion inNatural Env., ASTM STP 558, , 1974, pp.7-22

15- Griffin, R.B., Marine Atmospheres, Metals Handbook, Ninth Edition,Volume 12, ASM, 1987, pp. 902-906.

16- Section 09874 Specifications for Protective Coating and Lining System, TheRoyal Commission for Jubail and Yanbu, Feb. 2000.

17- Lendiva-Linter, E, R. J. Franco and J.S. Beecher, An Innovative Approach toControlling Salt Water Cooling Tower System Problems, Paper Number 85,CORROSION/78, NACE, 1978.

18- Engineering New Record (ENR), Cover Story PETROLEUM Giant NitrogenPlant Enhances Oil Production in Mexico, pp. 32-35, May 8, 2000 Issue.

Seawater Feasibility

Documents

Transcript of Seawater Feasibility